The present invention generally relates to a material for and method of joining metals and, more particularly, to improved materials and methods for brazing titanium alloys including Beta 21S.
Titanium alloys have been of considerable interest in many applications due to their desirable performance characteristics. Such alloys provide low density, high strength, fatigue resistance, corrosion resistance, and good strength-to-weight ratio. Because titanium alloys weigh a great deal less than stainless steel and nickel based alloys, they have provided great advances in many fields where there is a constant drive to minimize design weight. For instance in aerospace applications, aircraft heat exchangers and the like greatly benefit from reduced weight. The operating conditions of heat exchangers also involve high stresses induced by pressure and temperature, combined with fatigue loading. Another desirable characteristic of titanium alloys is their ability to withstand high temperatures, as the temperature in aircraft heat exchangers can be in excess of 1000° F.
Of the various titanium alloys that exist, metastable beta-titanium alloys are of great interest, particularly in aerospace applications requiring highly formable sheet metal or foil gages. One of the most promising beta alloys is Beta 21S, i.e., a beta alloy containing about 21% of alloying additions. Beta 21S, Ti-15Mo-3Nb-3Al-0.2Si, was developed to overcome some of the disadvantages of the other titanium alloys. As an example, alpha-beta alloys tend to have poor formability, while other beta alloys tend to have reduced elevated temperature properties. In contrast, Beta 21S has good formability, good elevated temperature properties, low density, and oxidation resistance.
Yet, the ability to employ Beta 21S in aerospace or other applications is limited by the ability to join pieces of Beta 21S together. Without the ability to adequately join, any application is limited in size and complexity. That is, the application is limited by the ability to make and form a single piece of a base material large enough to make the final product. If a mechanical joining process is needed to join multiple pieces of base material, weight savings from the base material itself may be lost. And the product design may require changes to accommodate a mechanical fastener. Additionally, the ability to attach objects to the base material can become limited by the physical presence of a mechanical fastener which might need to be located at the joiner point between the base material and the object.
On the other hand, the advantages of a non-mechanical joining process of base materials can be significant. Some non-mechanical or metallurgical joining processes have included welding, diffusion bonding and brazing. The advantages of non-mechanical joining can be most evident particularly when the base material is of a thin gage type and, thus, weight savings are increased. A thin gage material might be of an order around 0.002 to 0.090 inches thick. Also, and unlike a mechanical fastener, a non-mechanical joint can minimize the disadvantages of joining an object where the base material is itself joined. This minimization is achieved since the bulk or space occupied by mechanical fasteners are omitted.
Still, there are disadvantages from non-mechanical joining. The disadvantages can include excessive alloying, metallurgical interactions, dissolution and erosion of base materials, and degradation of mechanical properties. In spite of the disadvantages and because of the advantages provided by titanium alloys, including Beta 21S, considerable effort has been made in the past to improve non-mechanical joining. Much of the effort has recently focused on brazing.
Brazing may be generally characterized as exposing the base material and braze material to a temperature sufficient to cause the braze material to melt. The atoms from the braze material then interdiffuse with the atoms in the base material. Upon the braze material solidifying, a joint is formed. While the general brazing process appears to be clear and straightforward in principle, research indicates to the contrary. The quality of the braze joint is highly dependent upon various factors in the brazing process, such as temperature, rate of heating and cooling, composition of the braze material and composition of the base material. While the attempts to determine the effect of these factors have been many, their interdependent relationships remain less clear.
As an example of temperature and braze material dependency, C. Cadden et al., “Microstructural Evolution and Mechanical Properties and Braze Joints in Ti-13.4 Al-21.2 Nb,” Welding Research Supplement, pp. 316-325s (August 1997) addressed an alpha-two Ti base material with a Ti—Cu—Ni braze material. Cadden et al. indicate that, depending upon the braze temperature, the braze joint can have a room temperature tensile strength comparable to alpha-two and an elevated temperature (649° C. and 760° C.) tensile strength of 70 to 80% of the base metal tensile strength. Even with different nickel contents in the braze material, the average nickel content in the joint was found to be nearly constant. However, as between a rapidly solidified melt-spun braze foil and a laminated braze foil, the latter produced higher levels of nickel in the centerline of the joint, which was believed to lead to poorer room temperature tensile behavior. This indicates that the homogeneous rapidly solidified braze may have advantages over the braze comprised of foils (Cu/Ti/Ni) laminated together. However, melt spinning is an expensive process and cannot produce foil. Rather, melt spinning only produces narrow strips of varying thickness, which are difficult to use. Planner Flow Casting was developed to produce braze foil with a rapid solidified or amorphous structure. However, for Ti and Zr alloys, the nozzles used to cast the liquid metal and control its dimensions are quickly eroded and economic production of the desired braze foils is not possible.
In another study of how temperature can affect the braze joint, T. Onzawa et al., “Brazing of Titanium Using Low-Melting Point Ti-Based Filler Metals,” Welding Research Supplement, pp. 462-467s (December 1990) investigated the base materials of commercially pure titanium (CPTi) and Ti-6 Al-4 V. The different filler metals used with the base materials included Ti-37.5Zr-15Cu-10Ni, Ti-35Zr-15Cu-15Ni and Ti-25Zr-50Cu. Onzawa et al. concluded that brazing above the alpha-beta transformation temperature and the beta transus temperature of the base metal would cause the grains in the base metal to coarsen and fine Widmanstatten structure to form at the joint area. This resulted in poor mechanical properties. Below the transition temperatures, the fine grains of the base metals were preserved, as well as the braze zone being distinct from the braze metal. This led to better mechanical properties. Onzawa et al. also determined that a shorter holding time at a braze temperature could improve the mechanical properties.
The effect of temperature and cooling rate on Beta 21S was described by Huang et al., “Effect of Heat Treatment on the Microstructure of a Metastable .beta.-Titanium Alloy,” Journal of Materials Engineering and Performance, v. 3(4), pp. 560-566 (August 1994). They found that alpha precipitated preferentially on the grain boundaries during higher temperature aging and within the grains during lower temperature aging. High temperature solutioning produced a coarse grain size, while resolutioning treatment followed by slow cooling (such as during brazing) resulted in alpha precipitation. But with air cooling, precipitation was suppressed.
Another temperature and cooling rate investigation involved Ti—Pd and Ti-6Al-4 V alloys brazed with 25Ti-25Zr-50Cu. Botstein et al. “Brazing of titanium-based alloys with amorphous 25 wt. % Ti-25 wt. % Zr-50 wt. % Cu filler metal,” Materials Science and Engineering, pp. 305-315 (1994). Botstein et al. determined that high heating and high cooling rate created only traces of Widmanstatten structure at the joint interface. On the other hand, low heating and low cooling rate tended to result in a coarse dendritic structure having high microhardness and low fracture strength.
In investigating the confluence of temperature, cooling rate, braze material and base material, Rabinkin, “New Applications for Rapidly Solidified Brazing Foils,” Welding Journal, pp. 39-46 (October 1989) described rapid solidification as a process having high cooling rates that allow stabilization of alloys into an amorphous state. Because such alloys provide “instant melting,” Rabinkin indicated that they can be used to braze at lower temperatures and for a shorter time. As pointed out, these features are well suited to brazing items such as fine-gauge honeycomb which require protection from molten filler metals. More importantly, according to Rabinkin, is the ductility of the alloys, thus eliminating the need for large joint clearances to fill the braze cross-section. Rapidly solidified foils of 75Ti-15Cu-15Ni (sic) and 83.1Zr-16.9Ni were considered by Rabinkin to be advantageous for titanium base materials. On the other hand, Rabinkin indicated that braze powders have drawbacks that the rapidly solidified foils overcome. Apparently included in the group of disadvantageous powders is Ti—Zr—Cu—Ni which had been used on titanium based alloys, including tubing and honeycomb aircraft structures. However, we have already indicated that the plannar flow casting technique used to produce these amorphous foils relies upon a nozzle that is quickly eroded by molten Ti and Zr, resulting in short production times and high costs. There are also problems associated with casting the Ti in air, as usual occurs with planar flow casting. Casting in vacuum or inert gas is possible, but significantly increases costs and makes the production of foil with a uniform thickness and width even more difficult. A variation in foil thickness will affect the ability of the foil to produce a good braze, as the gap between the two parts to be brazed varies. For example, in some areas the braze may not touch the two parts to be brazed, and therefore may not braze them both. Also, variations in width require the foil to be hand cut and positioned. This also results in increased costs.
U.S. Pat. No. 6,149,051, issued to (AlliedSignal) Honeywell discloses a method of brazing a Ti-15 Mo-3 Nb-3 Al-0.2 Si base material including the steps of coating a braze material onto a base material. The braze material comprises substantially only a Ti—Cu—Ni—Zr mixture, with the mixture comprising about 40% Ti, 20% Cu, 20% Ni and 20% Zr. The braze material is heated then a braze joint is formed between the braze and base materials. The heating step can occur from 760 to 932° C. and over 15 to 90 minutes. While this solves many of the problems of incomplete brazing, it is a very costly method especially because of the addition of the extremely expensive Zr, but also because the techniques to make the braze alloy, such as powder atomization, mechanical alloying or mechanically commutation (i.e. breaking up a large ingot to powder), are expensive. Also, the brazing must be closely controlled requiring an initial heating and a subsequent heating. The '051 patent briefly states that as well as an alloyed powder the braze composition might be obtained by admixing elemental powders, but gives no details. This is because while this approach is recognized as feasible it has not been considered desirable. Problems with it are that the braze is very non-homogeneous and therefore melts over a wide temperature range making brazing difficult and producing a poor low strength braze joint. We have already discussed the problems described by Rabinkin “New Applications for Rapidly Solidified Brazing Foils,” Welding Journal, pp. 39-46 (October 1989) for alloyed powders. These are significantly increased for elemental admixed powders, which would be even more “disadvantageous” than prealloyed powders. Another problem is that these elemental powders have high oxide contents, which also reduces the properties of the braze. It would be desirable to use these admixed elemental powders in order to reduce cost, but this would require the disadvantages of a wide melting range, non-homogeneous braze microstructure and composition and the high oxide problem to be solved.
Further, it would be desirable to have a braze that allows for a second brazing at a lower temperature. For example, a heat exchanger may have 1000 joints that need to be brazed. The first braze may only effectively braze 90% of the joints, leaving 10% or 100 joints with leaks. It would be desirable to go back and rebraze the 100 joints that did not properly braze, rather than scrap the part. However, not at the risk of harming the 90% of the brazes that properly brazed. The present invention solves this problem, by providing a first brazing and optionally a second brazing with a lower melting temperature, yet at the same time no or minimal amounts of Zr.
Further, it would be desirable to decrease the thermal exposure experienced by the titanium parts being brazed. The cited prior works describe rapid heating and cooling cycles and the short times at temperature. A reduction in the braze temperature, as we show to be possible, would greatly reduce thermal exposure.
As can be seen, there is a need for improved methods for brazing beta titanium alloys. There is a particular need for improved brazing methods that are less temperature and/or time dependent such that processing parameters need not be so tightly controlled, and there is also a related need to reduce braze temperature. A further need is a braze which is does not erode the titanium substrate. This is particularly important for thin foil, as any erosion will quickly melt through the foil. Erosion occurs due to the required use of braze temperatures higher than the melt temperature of the braze material. Also, erosion occurs because the melt temperature of the braze changes as Ti from the substrate is dissolved into the braze. This in turn changes its composition, i.e. the use of a non-optimum braze composition. There is also a need to produce a homogeneous braze microstructure, without the expense of rapid solidification techniques. Likewise, there is a need to provide an improved brazing method, which enables other objects to be welded at the brazed joint to further fabricate a complex assembly, but without deteriorating the brazed joint. There is a further need to provide a brazing method which can utilize a powder braze material which is easy and economical to formulate. Also, there is a need for a braze material that allows for a second brazing, when needed, at a lower temperature.